Instructions to use audarai/Audar-ASR-V1-Flash with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use audarai/Audar-ASR-V1-Flash with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("automatic-speech-recognition", model="audarai/Audar-ASR-V1-Flash", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("audarai/Audar-ASR-V1-Flash", trust_remote_code=True)
model = AutoModelForMultimodalLM.from_pretrained("audarai/Audar-ASR-V1-Flash", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = processor.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

llama-cpp-python

How to use audarai/Audar-ASR-V1-Flash with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="audarai/Audar-ASR-V1-Flash",
	filename="Audar-ASR-V1-Flash-Q4_K_M.gguf",
)

llm.create_chat_completion(
	messages = "\"sample1.flac\""
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use audarai/Audar-ASR-V1-Flash with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf audarai/Audar-ASR-V1-Flash:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf audarai/Audar-ASR-V1-Flash:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf audarai/Audar-ASR-V1-Flash:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf audarai/Audar-ASR-V1-Flash:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf audarai/Audar-ASR-V1-Flash:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf audarai/Audar-ASR-V1-Flash:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf audarai/Audar-ASR-V1-Flash:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf audarai/Audar-ASR-V1-Flash:Q4_K_M

Use Docker

docker model run hf.co/audarai/Audar-ASR-V1-Flash:Q4_K_M

LM Studio
Jan
Ollama
How to use audarai/Audar-ASR-V1-Flash with Ollama:
```
ollama run hf.co/audarai/Audar-ASR-V1-Flash:Q4_K_M
```

Unsloth Studio

How to use audarai/Audar-ASR-V1-Flash with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for audarai/Audar-ASR-V1-Flash to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for audarai/Audar-ASR-V1-Flash to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for audarai/Audar-ASR-V1-Flash to start chatting

Atomic Chat new
Docker Model Runner
How to use audarai/Audar-ASR-V1-Flash with Docker Model Runner:
```
docker model run hf.co/audarai/Audar-ASR-V1-Flash:Q4_K_M
```

Lemonade

How to use audarai/Audar-ASR-V1-Flash with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull audarai/Audar-ASR-V1-Flash:Q4_K_M

Run and chat with the model

lemonade run user.Audar-ASR-V1-Flash-Q4_K_M

List all available models

lemonade list

Audar-ASR-V1-Flash · Transformers + GGUF

Audar's proprietary Arabic ASR — the real-time, edge tier.

From Arabic to the world.

🧭 Overview · 📊 Benchmarks · 🤗 Transformers · 💻 GGUF · 🎙️ Streaming · ☁️ Audar API · 📜 License

🧭 What it is

Audar-ASR-V1-Flash is the edge tier of Audar's proprietary Arabic speech-recognition family — the same in-house Arabic training program as Audar-ASR-V1-Turbo, delivered in a fast ~0.6B-decoder model for real-time captioning and on-device use. It recasts transcription as audio-conditioned next-token prediction (a language-model decoder, not CTC/transducer), and is developed through Audar's proprietary pipeline:

🧱 Large-scale dialectal pretraining — 300,000+ hours of Arabic audio (MSA + Gulf, Egyptian, Levantine, Maghrebi; code-switching; diverse channels).
🎯 Dialect-targeted fine-tuning with hardness and multi-task sampling.
🧠 GRPO reinforcement-learning alignment from trained native-Arabic annotators.

It transcribes MSA and every major Arabic dialect, code-switched Arabic–English, and English, across 30 languages, and runs on CPU / GPU / edge via 🤗 Transformers or GGUF. For maximum accuracy on the hardest dialectal audio, use the larger Turbo tier.

Distributed in the widely-supported Qwen3-ASR architecture format for turnkey tooling (Transformers, llama.cpp / GGUF). The model — data, training curriculum, and alignment — is Audar's.

Model summary

Model	Audar-ASR-V1-Flash — proprietary Arabic ASR (edge tier)
Task	Automatic speech recognition (audio → text)
Approach	Generative ASR — audio encoder + language-model decoder
Training	300k+ hrs dialectal pretraining → dialect-targeted SFT → GRPO alignment
Decoder size	~0.6B (full model ~0.78B, bf16)
Audio input	16 kHz mono; 30 s context (longer audio is chunked/streamed)
Languages	Arabic (MSA + Gulf/Egyptian/Levantine/Maghrebi dialects) + English + 28 more
Runtimes	🤗 Transformers (GPU) · GGUF / llama.cpp (CPU · GPU · edge)
License	AudarAI Open License v1.0

📊 Benchmarks

Audar-ASR sets the state of the art on dialectal Arabic. The fully-benchmarked Turbo tier posts the lowest average WER of any evaluated system on the Open Universal Arabic ASR Leaderboard (24.7 % on the full test sets, best on four of six), 3.55 % WER on CommonVoice-18 Arabic, and 19.4 % WER / 7.3 % CER on Emirati (Mixat) — see the Turbo card.

Flash is the edge tier: it trades some accuracy for latency and throughput. Measured with our harness on an internal dialectal validation sample (WER/CER %, N clips per set):

Set (dialect)	N	WER %	CER %
SawtArabi (Gulf)	23	37.4	16.0
ArzEn (Egyptian ⇄ English code-switch)	40	43.7	19.5
MGB-3 (Egyptian broadcast)	40	49.0	18.1
Casablanca (Maghrebi / Moroccan Darija)	40	71.3	28.7

Use Flash for real-time and on-device transcription; step up to Turbo when you need the lowest error on heavy dialectal or long-form audio (on this same sample, Turbo roughly halves Flash's WER: Gulf 37→14, Egyptian code-switch 44→20).

🤗 Transformers inference

Ships self-contained modeling code, so trust_remote_code=True is required.

# pip install "transformers>=4.57" torch librosa
import re, torch, librosa
from transformers import AutoProcessor, AutoModelForCausalLM

repo = "audarai/Audar-ASR-V1-Flash"
proc  = AutoProcessor.from_pretrained(repo, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    repo, trust_remote_code=True, torch_dtype=torch.bfloat16, device_map="cuda:0",
).eval()

SYSTEM = "فرّغ الكلام العربي التالي."          # "Transcribe the following Arabic speech."
audio, _ = librosa.load("clip.wav", sr=16000, mono=True)

conv = [
    {"role": "system", "content": SYSTEM},
    {"role": "user",   "content": [{"type": "audio"}]},   # audio placeholder (a list, not "<audio>")
]
text   = proc.apply_chat_template(conv, tokenize=False, add_generation_prompt=True)
inputs = proc(text=text, audio=audio, sampling_rate=16000, return_tensors="pt").to(model.device)
inputs["input_features"] = inputs["input_features"].to(model.dtype)   # features are fp32 → cast to bf16

out = model.generate(**inputs, max_new_tokens=440, do_sample=False)
hyp = proc.batch_decode(out[:, inputs["input_ids"].shape[1]:], skip_special_tokens=True)[0]
print(re.sub(r"^\s*language\s+[A-Za-z]+\s*(?:<asr_text>)?\s*", "", hyp).strip())

Language steering: the Arabic auto-dialect prompt above needs no dialect hint. For other languages use e.g. "Transcribe the following speech.".
Long audio (>30 s): split at ~30 s boundaries (see the streaming section).

💻 GGUF inference (llama.cpp)

Audar-ASR runs on llama.cpp via the multimodal (mtmd) path: a quantized decoder GGUF plus a BF16 audio projector (mmproj). Build a recent llama.cpp (with Qwen3-ASR support), then:

./llama-mtmd-cli \
  -m       Audar-ASR-V1-Flash-Q8_0.gguf \
  --mmproj mmproj-Audar-ASR-V1-Flash.gguf \
  --audio  clip.wav \
  -sys     "فرّغ الكلام العربي التالي." \
  --temp 0

⚠️ The audio projector (mmproj) must stay BF16 — the encoder's ClippableLinear is numerically sensitive, so F16/Q8 measurably degrade quality. The decoder quantizes normally.

GGUF variants

File	Approx. size	Notes
`Audar-ASR-V1-Flash-Q4_K_M.gguf`	~0.40 GB	Smallest; best for edge/offline
`Audar-ASR-V1-Flash-Q8_0.gguf`	~0.64 GB	Near-lossless, CPU-friendly (recommended)
`Audar-ASR-V1-Flash.gguf` (BF16)	~1.20 GB	Full precision decoder
`mmproj-Audar-ASR-V1-Flash.gguf`	~0.38 GB	BF16 audio encoder — required, keep BF16

Prefer a managed endpoint? The Audar-ASR family is also available via the Audar API/SDK — streaming, speaker-attributed transcription, and diarization, production-hosted.

🎙️ Real-time streaming

The 30 s-context model streams via LocalAgreement-2: as audio arrives, the trailing window is re-decoded each hop and a word is committed only once two consecutive decodes agree on it — giving stable, low-latency incremental output on both the Transformers and GGUF paths. Audar's production realtime engine serves the same policy over an OpenAI-Realtime-compatible WebSocket with model-based endpointing.

🌍 Languages, dialects & tasks

Primary: Arabic — MSA and dialectal (Gulf/Emirati, Egyptian, Levantine, Maghrebi), plus code-switched Arabic–English; dialect-faithful orthography from audio alone.
Also: English + 28 additional languages.
Task: transcription (audio → UTF-8 text), prompt-steerable for language/formatting.

Intended use & limitations

Intended use. Live captioning and subtitles, voice assistants/agents, meeting and call-center transcription, media/broadcast, accessibility — cloud, on-prem, or offline/edge.

Limitations.

Maghrebi / Moroccan Darija (Casablanca) is the hardest condition for all systems.
Heavily code-switched telephony and low-SNR audio degrade accuracy relative to clean MSA.
Long recordings can drift; chunk at sentence boundaries for best results.
Not evaluated for, and must not be used for, covert speaker identification.

📜 License

Released under the AudarAI Open License v1.0 — commercial use, redistribution, and fine-tuning/quantization permitted; ship the license and keep notices. See audarai.com/license/audarai-open-license-v1.0.

Citation

@misc{audar-asr-flash-2026,
  title  = {Audar-ASR: Dialect-Aware Arabic Speech Recognition},
  author = {AudarAI},
  year   = {2026},
  note   = {Audar-ASR-V1-Flash},
  url    = {https://huggingface.co/audarai/Audar-ASR-V1-Flash}
}

About AudarAI

Leading Arabic-First Multilingual Audio Intelligence

AudarAI starts with Arabic — and expands to the world.

We are building advanced multilingual audio intelligence that helps individuals, enterprises, and governments communicate across languages, cultures, and borders. By combining Arabic-first speech technology with global multilingual AI, AudarAI transforms voice into understanding, interaction, and connection.

Our work spans speech recognition, speech understanding, voice-enabled digital assistants, human-computer interaction, and intelligent audio systems designed for real-world impact. From empowering people to access technology in their native language to helping organizations communicate globally, AudarAI is shaping a future where every voice can be heard, understood, and connected.

Arabic-first. Multilingual by design. Human-centered at heart.

🌐 www.audarai.com · 🤗 Hugging Face · GitHub · contact@audarai.com

Downloads last month: 591

Safetensors

Model size

0.8B params

Tensor type

BF16